The reason these posts aren't being filled is that they're almost exclusively looking for AI researchers with postgraduate degrees. Admittedly, knowing the underlying math is incredibly useful if you're focusing on the core functionality of a library or trying to eke out that last bit of precision, but for a ton of applications, having some number sense and an understanding of core statistical/linear algebra principles is more than sufficient (i.e. you don't have to know how to solve by hand, just have an idea of what the numbers are supposed to look like and be able to tell when something's not right). You don't need a postgrad degree to do this, or even a Bachelor's (though you might have trouble convincing a company of this since they seem to be looking for PhDs). I expect that as need increases and more people of varying levels of education start messing with ML for their own projects, the system will move towards certifications, analogous to the cybersecurity industry.
/s
Now that that's out of the way, PHP is probably one of the great democratizing influences on the internet. Should it have been designed with security in mind from the get-go? Yes. Is it inherently bad? No, and there's a lot of cool stuff you can do with it. I think a more restrictive default configuration would go a long way (and also help to discourage poor coding practices), and a lot of issues have been addressed in the newer 7.x versions. There's probably good money to be made in converting PHP 5.x sites, and in most cases, it'll be an easier job than rewriting the whole site from scratch in {javascript flavor of the month} or.NET, not to mention probably more affordable to small businesses.
Good point. Slashdot needs to go all bumpstops, all the time. Lets cut out the silly news aggregation aspect of it and turn it into a single-topic soapbox, as it was originally intended to be.
Don't ask me. I'm still bitter they shut down my Basis Peak. Yet another awesome thing that Intel decided to pull the plug on after buying and dabbling with it for a little bit.
Something Something Network Investment. Something Something Title II. It's pretty hard to claim that you WANT to invest in infrastructure when you're suing to breach a contract that said you'd invest in infrastructure.
Are they ok with automatically generated and mailed reports, or do they expect you to manually predigest the information and draw conclusions for them (with pretty colors and maybe a cute animal picture at the bottom)? The first is a perfectly acceptable and legitimate request, the second, while extremely common, is not.
Funny, a lot of the caliphates in the history books largely thrived because they encouraged math and social sciences, not to mention chemisty and physics.
The major problem with online learning is that 99% of them are lectures. Even when I was in school, I'd just read the textbook during lectures; I learn better from textbooks than I do from lecture videos, and we all know how dry textbooks are. Now something interactive and engaging? That would be fun. Even break the material into bite-sized text chunks, or maybe infographics, with regular quizzes, and it would be better than the current "stare at a video of a guy lecturing for an hour".
This. Onboard is good enough for playing a game, even with 5.1. Add some nice flat-response studio monitors, and they'll hiss like crazy off the onboard sound card (yet sound wonderfully clean off of a decent tablet or smartphone). Build a digital audio workstation, and enjoy trying to get latency low enough to actually use realtime midi input. Solved all my problems with an old M-audio pci card for less than a hundred bucks. Plus it does balanced line-out. Now I don't hear the refrigerator kicking on over my speakers.
This already exists; the first 30% of any video can be skipped because it contains no interesting content. Indeed, you can append "&wadsworth=1" to the end of any YouTube video URL to load the video 30% in. No algorithm is really necessary, though I suppose wadsworth could be used as a starting point for a bidirectional search to speed up the algorithm if you really wanted to be sure.
I'm surprised no one (especially the guy who proposed it) has thought of this, but you wouldn't need a *wall*, per se. It would be MUCH cheaper to have a series of columns or earth mounds (or column reinforced earth mounds) spread out in the right locations (think pachinko array). Check out the seismic cloak and tsunami cloak ideas. You don't need to prevent the winds from mixing, just make sure they do it in a controlled manner so that the energy is dispersed more evenly. This would have the additional benefit of preserving most of the "normal" weather patterns in the area that allow crops to grow properly.
Since this is a meta-analysis, it doesn't take into account a lot of other factors; many people with mental illness come from or end up in lower socioeconomic places.
But more than anything else, people with mental illness SMOKE. Not all of them, and not all the time, but tobacco usage among the mentally ill is 70% higher according to the CDC than it is amongst everyone else. In this case the correlation is almost a doozy.
These also apparently charge more quickly, which is definitely on the list of "stuff we really want from batteries". This definitely seems to be at the "might actually go into production" phase, and if they're telling the truth it'll make for cheap production as well. I'll trade power density for cheap, fast charging, and reusable when it comes to anything I don't have to carry myself.
Data Science will be around for a while yet, but the truth is most organizations that hire expensive data scientists will collect data, analyze it, and proceed to ignore any recommendations and proceed on gut instinct, or overcorrect retroactively, rather than making decisions based on the analyzed data. Sometimes this works, sometimes it doesn't. Depends on whose gut instincts are being followed.
Understanding specific older technologies may not be necessary, but older technologies tend to offer a MUCH more intuitive grasp of first principles than most modern technology (with some exceptions).
It also teaches that technology doesn't succeed because it's "better", but because it scales cheaply and easily. The Romans knew how to make steel, but their bronze technology was advanced enough in comparison that their best swords were made out of bronze; they used steel for the cheap mass produced swords. Tube amps sound better than most solid state amps, but solid state amps are cheaper to produce and maintain, so they dominate the market.
Additionally, some knowledge that has been superseded may become useful in the future. Very few people have the kind of knowledge of analog electronics that was commonplace in engineering 50 years ago (slashdot is probably an exception), but if you want to create a solution that is optimal, rather than "good enough" (special projects, etc.), there are a lot of analog optimizations that can improve the performance of digital equipment (by lowering the noise floor and isolating ground loops, and far more esoteric things that I've never heard of).
Last of all, the current legal state of affairs is set to quash innovation except in very specific directions, and that combined with the pressing need for consistent standards means we'll likely see innovation in many fields grind to a near standstill, though of course new fields will open up.
TL;DR technology is not a march forward so much as a drunken, weaving stagger, and understanding it and the forces that drive it provides a much more firm knowledge base than knowing how to work [tech du jour]. People with this type of knowledge also seem to pick up on how to work new technology faster, rather than being stuck with what they learned how to operate in their youth.
Slashdot Mirror
The reason these posts aren't being filled is that they're almost exclusively looking for AI researchers with postgraduate degrees. Admittedly, knowing the underlying math is incredibly useful if you're focusing on the core functionality of a library or trying to eke out that last bit of precision, but for a ton of applications, having some number sense and an understanding of core statistical/linear algebra principles is more than sufficient (i.e. you don't have to know how to solve by hand, just have an idea of what the numbers are supposed to look like and be able to tell when something's not right). You don't need a postgrad degree to do this, or even a Bachelor's (though you might have trouble convincing a company of this since they seem to be looking for PhDs). I expect that as need increases and more people of varying levels of education start messing with ML for their own projects, the system will move towards certifications, analogous to the cybersecurity industry.
/s Now that that's out of the way, PHP is probably one of the great democratizing influences on the internet. Should it have been designed with security in mind from the get-go? Yes. Is it inherently bad? No, and there's a lot of cool stuff you can do with it. I think a more restrictive default configuration would go a long way (and also help to discourage poor coding practices), and a lot of issues have been addressed in the newer 7.x versions. There's probably good money to be made in converting PHP 5.x sites, and in most cases, it'll be an easier job than rewriting the whole site from scratch in {javascript flavor of the month} or .NET, not to mention probably more affordable to small businesses.
What's the point of having an oppressed populace if you can't monetize their clicks?
Good point. Slashdot needs to go all bumpstops, all the time. Lets cut out the silly news aggregation aspect of it and turn it into a single-topic soapbox, as it was originally intended to be.
Don't ask me. I'm still bitter they shut down my Basis Peak. Yet another awesome thing that Intel decided to pull the plug on after buying and dabbling with it for a little bit.
Something Something Network Investment. Something Something Title II. It's pretty hard to claim that you WANT to invest in infrastructure when you're suing to breach a contract that said you'd invest in infrastructure.
aww crap someone beat me to it
I believe the appropriate phrase is "Yo Dawg"
Are they ok with automatically generated and mailed reports, or do they expect you to manually predigest the information and draw conclusions for them (with pretty colors and maybe a cute animal picture at the bottom)? The first is a perfectly acceptable and legitimate request, the second, while extremely common, is not.
Funny, a lot of the caliphates in the history books largely thrived because they encouraged math and social sciences, not to mention chemisty and physics.
The major problem with online learning is that 99% of them are lectures. Even when I was in school, I'd just read the textbook during lectures; I learn better from textbooks than I do from lecture videos, and we all know how dry textbooks are. Now something interactive and engaging? That would be fun. Even break the material into bite-sized text chunks, or maybe infographics, with regular quizzes, and it would be better than the current "stare at a video of a guy lecturing for an hour".
I feel like comcast would have a much easier time with the whole net neutrality debate if they explained that data caps were against divine law.
This. Onboard is good enough for playing a game, even with 5.1. Add some nice flat-response studio monitors, and they'll hiss like crazy off the onboard sound card (yet sound wonderfully clean off of a decent tablet or smartphone). Build a digital audio workstation, and enjoy trying to get latency low enough to actually use realtime midi input. Solved all my problems with an old M-audio pci card for less than a hundred bucks. Plus it does balanced line-out. Now I don't hear the refrigerator kicking on over my speakers.
This already exists; the first 30% of any video can be skipped because it contains no interesting content. Indeed, you can append "&wadsworth=1" to the end of any YouTube video URL to load the video 30% in. No algorithm is really necessary, though I suppose wadsworth could be used as a starting point for a bidirectional search to speed up the algorithm if you really wanted to be sure.
I'm surprised no one (especially the guy who proposed it) has thought of this, but you wouldn't need a *wall*, per se. It would be MUCH cheaper to have a series of columns or earth mounds (or column reinforced earth mounds) spread out in the right locations (think pachinko array). Check out the seismic cloak and tsunami cloak ideas. You don't need to prevent the winds from mixing, just make sure they do it in a controlled manner so that the energy is dispersed more evenly. This would have the additional benefit of preserving most of the "normal" weather patterns in the area that allow crops to grow properly.
Carbonized/epoxied bamboo would work just fine as a natural carbon fiber, methinks.
Since this is a meta-analysis, it doesn't take into account a lot of other factors; many people with mental illness come from or end up in lower socioeconomic places. But more than anything else, people with mental illness SMOKE. Not all of them, and not all the time, but tobacco usage among the mentally ill is 70% higher according to the CDC than it is amongst everyone else. In this case the correlation is almost a doozy.
These also apparently charge more quickly, which is definitely on the list of "stuff we really want from batteries". This definitely seems to be at the "might actually go into production" phase, and if they're telling the truth it'll make for cheap production as well. I'll trade power density for cheap, fast charging, and reusable when it comes to anything I don't have to carry myself.
Not only this, but memories can be modified when retrieved, so memories that are frequently retrieved are at greater risk of being corrupted.
VPN
Data Science will be around for a while yet, but the truth is most organizations that hire expensive data scientists will collect data, analyze it, and proceed to ignore any recommendations and proceed on gut instinct, or overcorrect retroactively, rather than making decisions based on the analyzed data. Sometimes this works, sometimes it doesn't. Depends on whose gut instincts are being followed.
With enough blocks this could conceivably happen...
Understanding specific older technologies may not be necessary, but older technologies tend to offer a MUCH more intuitive grasp of first principles than most modern technology (with some exceptions). It also teaches that technology doesn't succeed because it's "better", but because it scales cheaply and easily. The Romans knew how to make steel, but their bronze technology was advanced enough in comparison that their best swords were made out of bronze; they used steel for the cheap mass produced swords. Tube amps sound better than most solid state amps, but solid state amps are cheaper to produce and maintain, so they dominate the market. Additionally, some knowledge that has been superseded may become useful in the future. Very few people have the kind of knowledge of analog electronics that was commonplace in engineering 50 years ago (slashdot is probably an exception), but if you want to create a solution that is optimal, rather than "good enough" (special projects, etc.), there are a lot of analog optimizations that can improve the performance of digital equipment (by lowering the noise floor and isolating ground loops, and far more esoteric things that I've never heard of). Last of all, the current legal state of affairs is set to quash innovation except in very specific directions, and that combined with the pressing need for consistent standards means we'll likely see innovation in many fields grind to a near standstill, though of course new fields will open up. TL;DR technology is not a march forward so much as a drunken, weaving stagger, and understanding it and the forces that drive it provides a much more firm knowledge base than knowing how to work [tech du jour]. People with this type of knowledge also seem to pick up on how to work new technology faster, rather than being stuck with what they learned how to operate in their youth.
This could be REALLY useful for assembling small, delicate systems in coordination with, say, 3D printing.
love me some zelazny